Shape agent capabilities into your workflow.

docker-wsl-cleanup reports the Docker virtual disk and `docker system df`, guides a confirmed prune, stops Docker Desktop cleanly, shuts down WSL, and prints the appropriate VHDX compaction command for Windows Pro, Enterprise, or Home. It restarts Docker and reports the reclaimed space after the user completes the administrative step. The skill requires Windows 10 or 11 with Docker Desktop using WSL2 and an administrator PowerShell; it never self-elevates or force-kills Docker.

cc-provider-add configures MiMo, GLM, Kimi, Bedrock, or another Anthropic-compatible endpoint for Claude Code using a separate `CLAUDE_CONFIG_DIR` per provider. It creates protected secret files, model settings, onboarding metadata, a launcher, and a shell alias, so multiple providers can run in separate terminals without changing `~/.claude/settings.json`. The skill verifies the connection with a real API ping. It is designed for terminal users of Claude Code, and trusting a third-party endpoint means sending prompts to that provider.

blpapi-expert helps Claude write Bloomberg Desktop API code for reference, bulk, historical, intraday, subscription, derivatives, override, and error-handling workflows. It consults bundled request schemas and canonical examples instead of guessing field shapes, while keeping xbbg version differences visible. The scope is local Desktop API sessions attached to a Bloomberg Terminal; B-PIPE, Server API, Platform, EMSX, and Excel COM are excluded. Running the generated code requires a logged-in Terminal plus Bloomberg's SDK and xbbg.

best-aeo-skill audits a URL with 33 local evidence collectors, scores technical accessibility, content citability, structured data, and entity signals, and reports how the site may be cited by ChatGPT, Claude, Perplexity, Gemini, and other answer engines. Its fix commands can draft `llms.txt`, robots rules, JSON-LD, and content changes as a reviewable diff, while monitor mode tracks score regressions. The tool runs without an external API and does not push changes without confirmation.

cera-reasoning-harness places an AI project's goals and instructions beside the chat, links them to each message, and saves the project locally so a later session can resume. It keeps a history of the interaction and its decisions; the project states that saved files are local and encrypted. The app targets Windows 10 or 11 and needs network access plus an AI service connection, so verify those privacy claims in the implementation before using it for sensitive work.

Timeln's skills combine six memory-recall actions with a human-gated consulting workflow. Thinking-OS retrieves decisions, failures, shipped artifacts, and call material through Timeln MCP, while Consulting-OS moves from framing and solution options through gates, acceptance, commercial terms, packaging, integrity checks, red-team review, and pursuit. The skills require a Timeln account and MCP token for hosted memory, and they state that every answer cites a source or plainly reports no record; the podcast tool can use a local TTS engine.

qa403-ubuntu-install-claude-skill provides a repeatable script, install guide, and Claude Code walkthrough for running QuantAsylum's QA40x Windows audio-analyzer software on Ubuntu with Mono. It covers dependencies, udev and libusb rules, Wine extraction, rendering workarounds, and QA403 connection troubleshooting, with tested paths for Ubuntu 20.04 and 24.04. The scope includes QA402/QA403 hardware and excludes macOS, the QA40x V2 branch, and ALSA playback setup.

Dorian reviews a codebase across ten technical-debt categories, assigns severity tiers, and returns prioritized findings with path-and-line evidence. It then maps those findings one-to-one onto a single character that falls from guardian to gothic horror, producing a model-agnostic image prompt. Paths, names, secrets, and internal URLs are scrubbed before the prompt is shown.

font-matchmaker profiles a project through its domain, audience, stack, brand, and communication style, then evaluates five typography axes before recommending a font system. The result can cover headings, body text, and code, with alternatives for different trade-offs and a rationale tied to concrete type characteristics. It is aimed at developers who want a reasoned choice instead of a default font.

claude-code-qc compresses the staging, commit, and push sequence into the /qc command while matching the repository's existing commit style. It stages named files, keeps hooks and GPG signing enabled, and refuses amend, no-verify, blind add-all, force-push, and AI attribution patterns. The first run may need to bootstrap GitHub CLI authentication through a browser.

This migration skill parses .twb or .twbx workbooks and maps their data sources, calculations, parameters, filters, worksheets, and dashboard zones into dbt models and Lightdash YAML. It emits model, chart, dashboard, and migration-report files, while surfacing unsupported features before validation with Lightdash lint and chart smoke tests. It is for workbook migration, not for authoring new Lightdash content, administering Tableau Server, or handling Tableau Prep .tflx/.twfl flows.

qa-defect-report-writer converts anything from a one-line bug note to a wall of logs into a consistent defect report with reproduction steps, expected and actual results, severity, and priority. It separates multiple bugs and labels inferred or assumed fields so missing facts are visible instead of silently filled in. The Markdown output is formatted for issue trackers such as Confluence, Notion, GitHub, and Jira.

Schema Detector fetches a single URL, extracts JSON-LD, Microdata, and RDFa, infers the page type, and runs twelve schema checks plus malformed-JSON detection. It returns deterministic JSON with findings ranked from P0 to P3 and maps each issue to one of Ygramul's seven SEO levers. The skill is deliberately single-page and does not edit schema or execute client-side JavaScript.

AI Tutor structures learning around retrieval practice, spaced review, interleaving, elaborative questions, and adaptive difficulty rather than immediate answer delivery. Its learn, practice, review, quiz, and explain modes prompt the learner to recall first and offer only the hint needed to continue. It is plain Markdown that can be adapted to several LLM platforms; where file uploads are supported, the learning-science reference files can be added as well.

decide researches current industry practice for an explicit implementation question, then acts autonomously when evidence converges on a reversible choice. When sources disagree, it asks for a factual piece of project context instead of presenting an arbitrary preference poll. This repository is now a legacy mirror whose active development moved to Decision Kernel, and its workflow depends on external research tools.

oss-due-diligence guides a four-phase assessment of open-source repositories and third-party dependencies: identity and license checks, adversarial review, operating posture, and incident response. It adjusts emphasis for project archetypes and includes shell scripts for dependency hygiene, risky patterns, secrets, and release signals. The methodology is explicit about what was checked and avoids treating a limited review as proof that a dependency is safe.

perplexity-plugin teaches Claude Desktop and Cowork how to route fresh-information tasks across Perplexity's search, answer, reasoning, research, computation, history, and diagnostic tools. It also handles plan-gated features by falling back to free-tier search patterns when needed. The package does not install an MCP server: users must connect one themselves, and plan limits should be checked as they change.

compass audits a long Claude Code session along two axes: drift from the original task and rot in the codebase. It reads the transcript plus cached Git, test, lint, file-size, TODO, and boundary signals, then returns a SAFE, CONSIDER, or STOP recommendation with a next action. It does not run tests or research the web, supports Claude Code JSONL transcripts only, and is now a legacy mirror.